methods of rfid data processing in intelligent systems for the identification...

IV International Conference on "Information Technology and Nanotechnology" (ITNT-2018)

Methods of RFID data processing in intelligent systems for the

identification and movement control of industrial products

A V Astafiev1, A A Orlov

1, D P Popov

1 and M V Pshenichkin

1

1Vladimir State University, Gorkij str. 87, Vladimir, Russia, 600000

Abstract. The article describes the development and research of methods of RFID data

processing to build intelligent systems that provide timely and reliable automatic movement

control and identification of industrial products. Conducted and presented an analytical review

of Russian and foreign scientific-technical base for the development of methods and algorithms

for movement control systems. The structure of the hardware-software complex of the system

developed. The presented method of movement control products. Experimental studies of the developed system and methods.

1. Introduction

One of the most important elements of the quality management system of industrial production, whichlargely determines the efficiency of its functioning, is the mechanism of identification, which makes it

possible to ensure traceability of products throughout the technological production cycle. Traceability

in production helps to ensure compliance with government requirements and international qualitystandards, to perform a fast and targeted tracking of the entire technological cycle of manufacturing

products, which, in turn, minimizes financial consequences. Especially relevant is the question of

tracing products in the enterprise, if the production cycle consists of many stages, implemented in

large production areas. The organization of the mechanism for tracing products is possible byautomating the control of the movement of industrial products.

According to the GOST 18353-79 standard, there are 9 methods of non-destructive testing, of

which 2 were widely used in the MCS: radio frequency identification and technical vision [1]. The use of technical vision approaches is complicated by the need for graphic marking of proper quality, which

is difficult to realize in real production conditions and requires significant financial and human

resources. The use of radio frequency identification is less demanding in the process of marking products. Traffic control, based on radio frequency identification methods, is an advanced information

technology for the construction of warehouse accounting systems.

A large number of companies around the world are engaged in developing software and hardware

for movement control systems (MCS), but the tasks of developing and implementing automated product tracking systems in production are still unresolved. At present, there are still a number of

problems in industrial enterprises, the solution of which is not realized by means of modern MCS.

These include: - not all radio frequency markings may be in the field of view of equipment;

- a number of existing MCS can not be used in view of safety restrictions;

- not all MCS can ensure continuous and correct operation in large industrial areas;

- a number of modern MCS can work only in the presence of positioning systems, which areequipped with transporting devices;


- the presence of interference and signals that make identification difficult.The listed problems do not allow to organize automatic traffic control at all sections of the

production process. To solve these problems, it is necessary to develop new methods that allow more

efficient processing of RFID data. Thus, the development of new methods for processing RFID data for the construction of automatic

systems that provide an operative and reliable control over the movement of industrial products is

topical. The aim of the project is to develop and study methods for processing RFID data for the

construction of intelligent systems that provide prompt and reliable automatic traffic control and

identification of industrial products.

To achieve this goal, it is necessary to solve the following tasks: 1. Analytical review of the Russian and foreign scientific and technical base on development of

methods and algorithms of MCS.

2. Development of the structure of the hardware-software complex MCS.3. Development of an intelligent system and methods for controlling the movement of products.

4. Experimental research and modernization of the developed methods.

2. Analytical review of the Russian and foreign scientific and technical base on the development

of methods and algorithms MCSThe PCS software is developed by: PCT-Invent (Russia, Sakt-Petersburg), AiTiProekt (Russia,

Moscow), Impinj (USA, Seattle), Motorola (USA, Morrisville), Nordic ID (Finland, Salo), FEIG

(Germany, Weilburg) and many others. However, the main activity of these companies is thedevelopment and production of the hardware part of radio frequency identification systems and the

creation of specialized software for working with it. This implies that for the development of

automatic identification systems based on the proposed software and hardware, it is necessary tocollect a whole team of technical specialists to write the project "from scratch."

To quickly obtain real-time information on the location of the product and all its movements,

systems based on the principles of radio frequency identification of product labels are used. The basis

of the systems is the method of remote receipt of product data by transmitting a radio signal from the RFID tag located on its surface to the tags recognition devices (read) and then writing the information

received to the database of computers forming the computer network. Such world-famous companies

as Wal Mart, METRO Group, Gillette, Procter & Gamble, Tesco, Benetton and others demonstrate on their practical experience the benefits of using RFID technology in the organization of automated

warehouse accounting and cargo control.

The automatic identification system (AIS) of objects of rolling stock of railway transport (locomotives, cars, and also large-capacity containers) "Palma" is known, including a corporate

computer network, code onboard sensors and reading points equipped at reference points, (RFID tags)

the information on the location of the monitored object is automatically removed.

The disadvantage of this AAL Palma is the limited functionality of the system, which consists in the fact that information from the RFID tag is read only when the object passes the specially equipped

information reading point (reader). Further location of the facility, for example in the warehouse, is

not automatically recorded, which reduces the reliability and efficiency of using AIS when solving the problem of cargo storage.

The closest approach to the proposed approach is the JPL RFID system for warehouse accounting

of metal pipes, based on the marking with radio frequency labels of plugs installed at the ends of

pipes. This approach allows the inventory of stored products in an automated mode using hand-held portable readers.

The disadvantage of this AIS is that the movement of pipe products is produced by various means

of minor mechanization in such a way that the marking installed on the product in most cases does not fall within the field of view of the reading sensors, which makes it impossible to track the products in

automatic mode.

A common disadvantage for this kind of AIS is that for the organization of automatic traffic control, readers are installed permanently, and the movement of the tracked products occurs as it

Data Science A V Astafiev, A A Orlov, D P Popov and M V Pshenichkin

207


passes along this device. An obligatory requirement is that the marking is in line of sight for the reader, which is practically impossible with the use of such transport devices as bridge cranes, loaders,

stackers and other means of minor mechanization. The movement of products for this type of

transportation is of a stochastic nature, and the location of the cargo can close the direct visibility of the reader to the marking.

The scientific community is also working on the introduction of RFID technologies in various

spheres of human activity. A great contribution to the development of radio frequency identification technology and SKD in various spheres of life was made by Bondarevsky AS, Zolotov RV, Do Zuy

Nyat, Kamozin D.Yu., Manish B., Shahram M., Ke-Sheng Wang, Worapot Jakkhupan, Somjit Arch-

int, Yuefeng Li, Mahir Oner, Alp Ustundag, Aysenur Budak and many others. The latest works in this

area are [2-7]. The application of these knowledge-intensive technologies makes it possible to automate the

processes of controlling the movement of industrial products at enterprises of various spheres of life

and, ultimately, to improve the efficiency and reliability of the control of transportation and warehouse accounting of manufactured products.

However, they are not without flaws. The use of existing software and hardware solutions is more

focused on the organization of automated warehouse accounting and is less suitable for automating

traffic control. In confirmation of this at a number of industrial enterprises.

3. Development of the structure of the hardware and software complex MCS

The development of intellectual MCS is aimed at solving the task of organizing and providing control

over the movement of products through the territory of industrial enterprises, handling and cargotransportation, including warehousing of goods equipped with multi-code marking. Information on the

identification of products can be collected in the database of computers that are part of the computer

network that encompasses the points of the cargo transportation route and stores data on the origin anddestination of the products and on its location in warehouses.

The proposed MCS consists of 4 main modules: server, client, product marking and marking of the

storage area.

Marking of storage areas is performed using Bluetooth-labels, and the current location of the small-scale mechanization means is determined using the IBeacon technology.

Labelling of products at the enterprise is done by attaching an RFID tag to them. In the event that

the product contains components shielding the signal from the mark, the marking is applied to several sides of the product.

The client is a stand-alone device equipped with an RFID reader to collect data from RFID tags of

products, a Bluetooth adapter for determining the current location, a microcomputer for collecting, processing and sending telegrams to the server and batteries to provide autonomous operation.

The server is a computing device that receives telegrams from client devices, compares the

received data with information from the database and generates messages about movements

transmitted to the server of the plant's process control system. The structure of the hardware-software complex of the MCS being developed is shown in Figure 1.

The developed architecture is applicable to various means of minor mechanization. Figure 2 shows

an example of placement on a bridge crane. The proposed system is explained with the following drawing (figure 2), on which it is indicated:

1 - Bridge crane, for moving products;

2 - Bridge crane truck;

3 - Bridge crane beams; 4 - Movable cargo or article;

5 - Customer;

6 - Product identification; 7 - Warehouse area limiter;

8 - Marking of the storage area.


208


Figure 1. The structure of the hardware and software complex of the developed MCS.

Figure 2. General scheme of means of moving pipes.

4. Development of an algorithm for determining the threshold for clipping markers that are not

involved in the movementThe aim of the work is to develop an algorithm for filtering RFID tags, which makes it possible to

exclude marks that are not involved in the course of movement and to conduct a series of laboratory

tests of the developed system for automatic control over the movement of products along untyped

routes.A disturbance is an external perturbation acting in the transmission system and preventing proper

reception of signals. Sources of interference may be either outside or inside the transmission system

itself. If the interference is regular and known, then fighting it is not difficult. For example, the background of an alternating current can be eliminated by compensation; interference from a

particular radio station with a modulation spectrum of normal width is eliminated by an appropriate

filter. The struggle with random interference presents the greatest difficulty. There are many methods of detecting interference. The following were considered in the work:

1. The arithmetic mean.

2. Determination of emissions in statistics.

3. Determination of the optimum noise level in the flow of spectra.4. Level of significance.

Two shelves have been created, each of which is marked with an iBeacon-tag. Portable products

are marked with RFID tags. Moves three products 1, 3, 4 from the first rack to the second one. Items c labeled 2 and 5 are located between the shelves and are not subjected to transfer. Table 1 presents the

primary data.

During the move, 23 signal level values (RSSI) were received from all RFID tags. To determine the threshold value of p, which will allow us to weed out marks that are not involved

in the movement, four methods were tested (Figure 3 a-d).


209


Table 1. Results of displacement. Label 1 2 3 4 5

Readings 69 56 71 68 53

71 52 75 75 -

71 - 74 76 -

77 - 77 83 -

79 - 77 80 -

78 - 79 81 -

78 - 81 - -

Total 523 108 534 463 53

a) b)

c)

d) Figure 3. a) Allocation of useful data on the threshold P, based on the arithmetic mean, b)

Determination of useful data on the histogram, c) Determination of the threshold by significance level

p=0.10, d) Determination of emissions.

Laboratory investigations were carried out, during which it was established that the method based

on the level of significance was the most suitable method for testing the tested methods. The determination of noise by percentage of the area or the arithmetic mean does not always yield correct

results. The emission detection method can not be used for this task, since false signals are not

discarded, due to the small sample that is obtained in most cases.

5. Experimental resultsTo test the evaluation of the reliability of the results obtained, an experimental study was carried out.

The subject of the study is a system for automatic control over the movement of products. The object

is a product, an enterprise or a warehouse.


210


Laboratory experience was conducted with a prototype system in conditions close to real production. To the moving cart was attached a model of the part of the beam, with the reading and

processing device mounted on the side. Under the traverse on the cart is mounted a pallet for products

in the form of pipes. View of the laboratory setup is shown in figure 4.

Figure 4. View of the laboratory setup.

For the experiment was created two racks (c1, c2), each of which is labelled iBeacon-tag. The movement of products made from rack 1 to rack 2. During the movement of production readings from

their RFID-tags and iBeacon-tags of racks have been processed and are presented in tables (Table 2,

3). Table 2. Readings from iBeacon-tags of racks.

Time Rack 1 Rack 2

35:22:00 0 43

35:28:00 0 26

35:34:00 0 0

35:40:00 0 0

35:46:00 3 0

35:52:00 34 0

35:58:00 37 0

36:04:00 27 0

36:10:00 27 0 36:16:00 16 0

Table 3. Readings from RFID tags of products.

Time Product 1 Product 2 Product 3 Product 4

35:22:00 65 16 91 89

35:28:00 61 5 101 101

35:34:00 75 20 105 105

35:40:00 81 11 108 107

35:46:00 73 9 110 109

35:52:00 57 21 98 95

35:58:00 38 7 103 110

36:04:00 38 21 99 110

36:10:00 61 2 107 110 36:16:00 35 3 78 80

The duration of the experiment was 1 minute. During the experiment with iBeacon-tags of rack has

received 212 readings and from RFID tags of products - 2715. Graphic interpretation of the data

presented in figures 5, 6.


211


Figure 5. Readings from iBeacon-tags of racks.

Figure 6. Readings from RFID tags of products.

From the graph in figure 4 shows that at time 00:18:00 readings from iBeacon-tag of the first rack

were no longer received by the reader. At time 0:24:00 began to receive readings from iBeacon-tag of

the rack number 2. This fact indicates that there was a movement of the transport device from the area of the rack 1 to the area of the rack 2.

Figure 5 shows the readings from the RFID-tags of products, time-spaced with an interval of 6

seconds. During the experiment, 4 items were moved. According to the graph it is clear that a stable signal came from all 4 product tags. This allows us to say that during the movement of the conveying

device the products moved along with it.

In the course of the work, more than 150 experiments were conducted in the laboratory. Also, the installation was tested in an industrial plant. The results of the experiment showed the reliability of the

movement identifications in the amount of 97.3%. As a result, information was also collected that

allowed adjustments to the system to improve its efficiency.

6. ConclusionA system and method for automatic control of the movement of products based on RFID-identification

when moving hoisting-and-transport mechanisms is developed. Considered and analyzed the existing

systems of automatic control of the movement of products based on radio frequency identification.The main advantages and disadvantages are revealed. The structure and levels of the hardware and

software complex MCS are described, the methodology of its operation is described. Thanks to the

presented system and methodology, the problems of moving several units of products simultaneously

were solved, the accuracy of detail of the location information was high, the client-server approachand the movement history analysis algorithms were implemented, the problem of authenticity of

0

10

20

30

40

50

Co

un

t o

f re

adin

gs

Time

Rack 1

Rack 2

0

20

40

60

80

100

120

Co

un

t o

f re

adin

gs

Time

Product 1

Product 2

Product 3

Product 4


212


identification and the tracking process outside the route was solved. Laboratory investigations were conducted, during which it was possible to obtain a high degree of reliability of the identification of

displacements in 97.3%, proving the urgency of the developed system and methodology. These studies

bear a high scientific value for providing control over the movement of products.

7. References

[1] GOST 18353-79 Nondestructive check. Classification of types and methods

[2] Astafiev A V, Orlov A A and Popov D P 2017 Development the algorithm of positioning

industrial wares in-plant based on radio frequency identification for the products tracking

systems CEUR Workshop Proceeding 1901 23-27

[3] Astafiev A V, Orlov, A A and Privezencev D G 2016 Method of controlling the movement of

large metal products with the use of algorithms for localization and recognition of bar code

markings Dynamics of Systems, Mechanisms and Machines, Dynamics 7819046 DOI:

10.1109/Dynamics.2016.7818969

[4] Orlov A A, Provotorov A V and Astafiev A V 2016 Methods and algorithms of automated two-

stage visual recognition of metal-rolling billets Automation and Remote Control 1099-1105

DOI: 10.1134/S000511791606014X

[5] Zhiznyakov A L, Privezentsev D G and Zakharov A A 2015 Using fractal features of digital

images for the detection of surface defects Pattern Recognition and Image Analysis 25(1)

122-131

[6] Nhat D D 2015 Researches and application of RFID technology (radio frequency identification)

International Research Journal 5 34-37

[7] Kamozin D Y 2013 Comparison of the effectiveness of bar-code technology and RFID

technology's application in logistics processes Bulletin of Baikal State University 3 71-75

[8] Parikh D and Jancke G 2008 Localization and segmentation of a 2D high capacity color barcode

Proceeding of IEEE Workshop on Applications of Computer Vision 1-6 DOI:

10.1109/WACV.2008.4544033

[9] Kazanskiy N L and Popov S B 2012 The distributed vision system of the registration of therailway train Computer Optics 36(3) 419-428

[10] Morozov A A and Sushkova O S 2016 Analysis of real-time video images using the means of

the Actor Prologue Computer Optics 40(6) 947-957 DOI: 10.18287 / 2412-6179-2016-40-6-

947-957[11] Epifantsev B N, Pyatkov A A and Kopeikin S A 2016 Multisensory systems for monitoring

restricted areas: the capabilities of a video analytics channel for intrusion detection Computer

Optics 40(1) 121-129 DOI: 10.18287 / 2412-6179-2016-40-1-121-129


213

methods of rfid data processing in intelligent systems for the identification...

Documents